A Subnational Age-Structured Mathematical Model of Malaria Transmission and Burden

Malaria transmission and burden exhibit substantial spatial and age-related heterogeneity that is often obscured by nationallevel modeling approaches. In this study, we develop a spatially explicit, age-structured malaria transmission and burden model designed to operate at the subnational scale (Admin-1) while remaining fully consistent with national-level estimates. The model integrates monthly age cohorts (0-100 years), naturally acquired immunity, heterogeneous entomological inoculation rates (EIR), and key non-vaccine interventions, allowing a fine-scale representation of malaria dynamics across space, time, and age. Subnational EIRs are inferred from malaria prevalence in children aged 2-10 years, while age-specific clinical progression and mortality are modeled through immunity-dependent transitions. To reconcile model-based outputs with the reported malaria burden, a parsimonious rescaling framework is introduced, based on time-varying country-level-to-subnational ratios of estimated cases and deaths, ensuring consistency between subnational projections and national burden trajectories. Model calibration is performed using a joint root-mean-square error (RMSE) metric that pools discrepancies across cases and deaths across all administrative units and years. Applying the model to both high-and low-transmission settings (Benin, Burkina-Faso, Ethiopia, and Somalia) successfully reproduces observed national and subnational dynamics, while revealing a strong and persistent spatial heterogeneity within countries. The model captures marked geographical gradients in the EIR, and localized epidemic hotspots with consistently higher values (especially in low-transmission countries), as indicated by subnational assessments. The resulting framework provides a flexible platform for analyzing spatial heterogeneity in malaria transmission, age-specific burden, and intervention impact, and is well suited for prospective analyses, including the evaluation of vaccination strategies, targeted intervention deployment, and long-term burden projections. The explicit representation of monthly age cohorts enables direct subnational assessment of malaria vaccines targeting infants and young children, offering a robust decision-support tool for geographically targeted malaria control and elimination planning.